Don't drop tables on import and specify database #838
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #838 +/- ##
=======================================
Coverage 94.92% 94.92%
=======================================
Files 90 90
Lines 3841 3841
=======================================
Hits 3646 3646
Misses 195 195 ☔ View full report in Codecov by Sentry. |
WillNilges
changed the title
| # XXX (willnilges): Would it be better to use manage.py? | ||
| for ((i = num_tables - 1; i >= 0; i--)); | ||
| do | ||
| $DOCKER_PG_COMMAND -c "DROP TABLE IF EXISTS ${tables[i]} CASCADE" |
There was a problem hiding this comment.
So looking at it now it's coming back to me why this might be needed. I think this should acutally be DELETE FROM for the type of export I generated for Sam :(
The import script ones were exported directly from psql with a command that might have been lost to time. These ones were generated with datagrip, and apparently they don't include the table definitions, just the insert statements. We should tune this script to whatever mechanized export process we want going forward (since the import script is no more)
There was a problem hiding this comment.
Oh interesting, I didn't realize you had used a separate import mechanism. I agree that this process needs some work, but we should not rely on datagrip for this. My favorite thing to do would probably be to grab a backup from the S3 bucket, and write a script (perhaps a manage.py script) that could do the import/scramble/export in one shot.
|
This will require some noodling. Closing for now. |
No description provided.